Battery SOH estimation method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of battery health, and discloses a battery SOH estimation method and device, computer equipment and a storage medium, and the method comprises the steps: collecting battery operation data and environment parameters in real time; constructing an environmental stress factor model based on the environmental data and a preset weight coefficient; calculating an environmental stress factor value in real time based on the environmental stress factor model, and constructing a mapping function between the environmental stress factor value and the battery health state corresponding to the timestamp based on the battery operation data; and obtaining a real-time battery SOH estimation value based on the battery operation data, the environmental stress factor value and the mapping function. According to the method, the built environmental stress factor model and the nonlinear mapping function are utilized, online real-time estimation of the SOH under the influence of different environmental factors is considered, the aging acceleration effect of the external environment on the battery is effectively reflected, and the estimation accuracy of the SOH of the energy storage system under the atrocious weather condition is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of battery health, and specifically relates to a method, device, computer device and storage medium for estimating the State of Health (SOH) of a battery. Background Art

[0002] With the transformation of the global energy structure towards cleaner and more sustainable forms, energy storage systems are playing an increasingly important role in the power system. The State of Health (SOH) of the batteries in an energy storage system is a key indicator for evaluating the remaining life and performance degradation of the batteries. Traditional SOH estimation methods are mostly based on single models or data-driven algorithms of internal battery parameters (such as voltage, current, internal resistance), but do not fully incorporate the impact of environmental factors on battery aging. Under harsh climates (such as high temperatures in deserts, low temperatures in polar regions, coastal salt spray, high humidity and heavy rain), environmental conditions can significantly accelerate the performance decay of batteries, leading to an increase in the estimation error of traditional methods, and further triggering safety risks in energy storage systems. Relevant experiments have proven that the SOH estimation error in a laboratory environment is usually ≤ 3%, but under extreme climates (such as a daily temperature difference > 40°C, humidity > 90% RH), the error of traditional methods can reach 15% - 20%. At the same time, the passivation of the positive electrode material caused by salt spray corrosion will increase the annual capacity decay rate of the battery by 1.8 times.

[0003] It can be seen that the above environmental problems have a great impact on the health status of energy storage systems, and there is a problem of insufficient estimation of the health status (SOH) of energy storage systems under harsh climate conditions in the prior art. Summary of the Invention

[0004] In view of this, the present invention provides a method, device, computer device and storage medium for estimating the SOH of a battery to solve the problem of insufficient estimation of the health status (SOH) of energy storage systems under harsh climate conditions.

[0005] In a first aspect, the present invention provides a method for estimating the SOH of a battery, the method comprising:

[0006] Real-time collecting battery operation data and environmental parameters;

[0007] Constructing an environmental stress factor model based on environmental data and a preset weight coefficient;

[0008] Based on the environmental stress factor model, calculating the environmental stress factor value in real time, and constructing a mapping function between the environmental stress factor value and the battery health status at the corresponding timestamp based on the battery operation data;

[0009] Obtaining a real-time battery SOH estimated value based on the battery operation data, the environmental stress factor value and the mapping function.

[0010] A method for estimating the state of health (SOH) of a battery provided by the present invention can more comprehensively reflect the influence of the external environment of the battery by collecting environmental parameters such as temperature, humidity, salt mist concentration, and temperature change rate in real time. Compared with the traditional single data collection method, combined with the battery operation data, using the constructed Environmental Stress Factor (ESF) model and non-linear mapping function, it can estimate the state of health (SOH) of the battery online and in real time under the influence of different environmental factors, effectively reflecting the accelerating effect of the external environment on battery aging, improving the accuracy of estimating the state of health (SOH) of the energy storage system under harsh climate conditions, extending the battery life and reducing the maintenance cost, and solving the problem of insufficient estimation of the state of health of the energy storage system under harsh climate conditions.

[0011] In an alternative embodiment, the environmental parameters include temperature, relative humidity, salt mist concentration, and temperature change rate; the preset weight coefficients include temperature weight coefficient, relative humidity weight coefficient, salt mist corrosion weight coefficient, and temperature change rate weight coefficient;

[0012] Construct an environmental stress factor model based on the battery operation data, environmental data, and preset weight coefficients, including:

[0013] Calculate the product of the temperature and the temperature weight coefficient to obtain the temperature term, the product of the relative humidity and the relative humidity weight coefficient to obtain the humidity term, the product of the salt mist concentration and the salt mist corrosion weight coefficient to obtain the salt mist term, and the product of the temperature change rate and the temperature change rate weight coefficient to obtain the temperature change rate term;

[0014] Construct an environmental stress factor model based on the temperature term, humidity term, salt mist term, and temperature change rate term, and perform offline training on the preliminary environmental stress factor model using offline calibration. Correct the preliminary environmental stress factor model according to the actual battery aging data to obtain the final environmental stress factor model.

[0015] A method for estimating the state of health (SOH) of a battery provided by the present invention comprehensively incorporates environmental parameters such as temperature, relative humidity, salt mist concentration, and temperature change rate, and sets corresponding weight coefficients for each parameter. By calculating the product of each parameter and the weight coefficient respectively, independent temperature term, humidity term, salt mist term, and temperature change rate term are obtained. This method can accurately quantify the influence of different environmental factors on battery aging, better conform to the working conditions of the battery in a complex actual environment, avoid model deviation caused by incomplete consideration of environmental factors, and lay a foundation for accurately evaluating the degree of battery aging. Performing offline training on the preliminary environmental stress factor model using offline calibration and correcting the model according to the actual battery aging data can effectively improve the accuracy and reliability of the model. Through comparison and calibration with real data, the model can more truly reflect the relationship between environmental factors and battery aging, and avoid estimation errors caused by inaccurate model parameters.

[0016] In an alternative embodiment, the temperature weight coefficient, relative humidity weight coefficient, and salt spray corrosion weight coefficient are set in the following manner:

[0017] Orthogonal experimental design is adopted to conduct charge and discharge cycle tests on the battery under different temperatures, different humidities, and different salt spray degrees;

[0018] When the battery capacity meets the preset capacity threshold, record the battery capacity, internal resistance, and EIS spectrum during the charge and discharge cycle tests within the preset number of times, and calculate the reciprocal of the number of cycles required for the capacity to decay to the preset capacity threshold as the battery aging rate;

[0019] Calculate the proportion of the mean square deviation of the battery capacity, internal resistance, and EIS spectrum respectively as the weight benchmark, and calculate the initial values of the temperature weight coefficient, relative humidity weight coefficient, and salt spray corrosion weight coefficient based on the weight benchmark.

[0020] A battery SOH estimation method provided by the present invention adopts orthogonal experimental design to conduct charge and discharge cycle tests on the battery under different temperatures, humidities, and salt spray degrees. This experimental design method can comprehensively investigate the effects of multiple factors (temperature, humidity, salt spray concentration) and their interactions on the battery performance with fewer experimental times. When the battery capacity meets the preset capacity threshold, record multi-dimensional data such as the battery capacity, internal resistance, and EIS spectrum during the charge and discharge cycle tests within the preset number of times. The capacity reflects the energy storage ability of the battery, the internal resistance reflects the conductive performance inside the battery, and the EIS spectrum can deeply reveal the electrochemical process and interface characteristics inside the battery. By comprehensively analyzing these data, the aging state of the battery can be reflected from multiple angles, avoiding the one-sidedness of single-index analysis. Calculating the initial values of the weight coefficients of temperature, humidity, salt spray concentration, and temperature change rate with the proportion of the mean square deviation of experimental data as the weight benchmark is to determine the weights based on the data performance during the actual aging process of the battery, closely combining the actual working characteristics and aging laws of the battery, so that the weight coefficients can objectively reflect the contribution degree of each environmental factor to the battery aging.

[0021] In an alternative embodiment, calculating the environmental stress factor value in real time based on the environmental stress factor model includes:

[0022] Substitute the environmental parameters into the environmental stress factor model to obtain the environmental stress factor value corresponding to the time stamp.

[0023] In an alternative embodiment, constructing a mapping function between the environmental stress factor value and the battery health state corresponding to the time stamp based on the battery operation data includes:

[0024] Calculate the percentage of battery capacity decay consistent with the time stamp of the environmental stress factor value, and use the percentage of battery capacity decay as the battery health state;

[0025] Construct a mapping function between the environmental stress factor value and the battery health state at the corresponding timestamp based on battery operation data, environmental stress factor values, battery health state, and a preset neural network.

[0026] A battery SOH estimation method provided by the present invention uses the percentage of battery capacity decay consistent with the timestamp of the environmental stress factor value as the battery health state, accurately corresponding to the real-time health condition of the battery under specific environmental stresses. Constructing a mapping function based on battery operation data, environmental stress factor values, and battery health state integrates multi-source information such as the battery's own operating parameters, external environmental influencing factors, and actual health status. Using a preset neural network to construct the mapping function, the powerful non-linear fitting ability of the neural network can automatically learn and extract complex features and rules in the data, effectively handling the highly non-linear mapping relationship between the environmental stress factor value and the battery health state. Emphasizing the consistency between the environmental stress factor value and the battery health state timestamp ensures the accuracy of the logical relationship between the data.

[0027] In an optional implementation manner, constructing a mapping function between the environmental stress factor value and the battery health state at the corresponding timestamp based on battery operation data further includes:

[0028] Introduce an adaptive weight and dynamic correction mechanism into the mapping function. When training the mapping function, use the mean square error as the loss function of the mapping function, and update the parameters of the mapping function through an adaptive optimization algorithm until the training error converges.

[0029] A battery SOH estimation method provided by the present invention introduces an adaptive weight mechanism, enabling the mapping function to automatically adjust the influence weights of various factors on the battery health state evaluation according to the changes in the input data. The addition of the dynamic correction mechanism allows the mapping function to continuously optimize itself as the battery ages and the environment changes. Using the mean square error as the loss function can intuitively measure the difference between the model prediction value and the actual battery health state, providing a clear target orientation for model optimization. The adaptive weight, dynamic correction mechanism, mean square error loss function, and adaptive optimization algorithm cooperate with each other to form an organic whole, comprehensively enhancing the comprehensive performance of the mapping function, enabling it to stably and accurately realize the mapping between the environmental stress factor value and the battery health state in complex environments and the dynamic process of battery aging.

[0030] In an optional implementation manner, obtaining the real-time battery SOH estimation value based on battery operation data, environmental stress factor values, and the mapping function includes:

[0031] Input the battery operation data and environmental stress factor values into the mapping function to obtain a preliminary battery SOH estimation value;

[0032] Perform online filtering and adaptive correction on the preliminary estimated value of the battery SOH to obtain the real-time estimated value of the battery SOH.

[0033] A method for estimating the battery SOH provided by the present invention inputs the battery operation data and the environmental stress factor value into the mapping function, fully integrating the battery's own operation state and external environmental influencing factors, and leveraging the powerful non-linear fitting ability of the mapping function. Online filtering is performed on the preliminary estimated value of the battery SOH to process the random noise and abnormal fluctuations in the input data in real time. The adaptive correction mechanism can dynamically adjust the estimation result according to the battery operation conditions and environmental changes, enabling the real-time estimated value of the battery SOH to quickly respond to the changes in the battery state. Even under complex and changeable working conditions, it can maintain high accuracy and adaptability. Second, the present invention provides a device for estimating the battery SOH, and the device includes:

[0034] A data acquisition module for real-time acquisition of battery operation data and environmental parameters;

[0035] An environmental stress factor model construction module for constructing an environmental stress factor model based on environmental data and preset weight coefficients;

[0036] A mapping function construction module for calculating the environmental stress factor value in real time based on the environmental stress factor model and constructing a mapping function between the environmental stress factor value and the battery health state at the corresponding timestamp based on the battery operation data;

[0037] A real-time battery SOH estimated value calculation module for obtaining the real-time estimated value of the battery SOH based on the battery operation data, the environmental stress factor value, and the mapping function.

[0038] Third, the present invention provides a computer device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the method for estimating the battery SOH according to the first aspect or any corresponding embodiment thereof.

[0039] Fourth, the present invention provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to cause a computer to execute the method for estimating the battery SOH according to the first aspect or any corresponding embodiment thereof.

[0040] Fifth, the present invention provides a computer program product, including computer instructions, and the computer instructions are used to cause a computer to execute the method for estimating the battery SOH according to the first aspect or any corresponding embodiment thereof. Description of the Drawings

[0041] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0042] Figure 1 is a schematic flowchart of a battery SOH estimation method according to an embodiment of the present invention;

[0043] Figure 2 is a schematic flowchart of another battery SOH estimation method according to an embodiment of the present invention;

[0044] Figure 3 is a schematic flowchart of yet another battery SOH estimation method according to an embodiment of the present invention;

[0045] Figure 4 is a schematic flowchart of still another battery SOH estimation method according to an embodiment of the present invention;

[0046] Figure 5 is a structural block diagram of a battery SOH estimation device according to an embodiment of the present invention;

[0047] Figure 6 is a schematic hardware structure diagram of a computer device according to an embodiment of the present invention. Specific Embodiments

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0049] The main defects of the prior art include: poor environmental adaptability, no dynamic mapping relationship between environmental parameters (such as temperature, humidity, salt fog concentration, etc.) and SOH degradation is established, and the influence of environmental factors on the accuracy of battery SOH estimation is not considered. The embodiments of the present invention provide a battery SOH estimation method, which constructs a mapping function between the battery SOH based on the Environmental Stress Factor (ESF), achieving the effect of accurately evaluating the health state of the energy storage battery considering environmental factors in a complex and changeable climate environment.

[0050] According to an embodiment of the present invention, an embodiment of a method for estimating the state of health (SOH) of a battery is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0051] In this embodiment, a method for estimating the SOH of a battery is provided, which can be used in the above energy storage system. Figure 1 It is a flowchart of the method for estimating the SOH of a battery according to an embodiment of the present invention, as Figure 1 shown, this process includes the following steps:

[0052] Step S101, collect battery operation data and environmental parameters in real time.

[0053] Specifically, a weather-resistant sensor network is arranged in the energy storage system to collect battery operation data (such as voltage, current, charge and discharge cycle times, etc.) and environmental parameters (such as temperature, relative humidity, salt mist concentration, temperature change rate, etc.) in real time. The weather-resistant sensor network includes a temperature and humidity sensor, a salt mist concentration sensor, a voltage sensor and a current sensor. The collected battery operation data and environmental parameters are also preprocessed, including data cleaning, noise filtering, interpolation and normalization processing, to ensure the accuracy and stability of the input data.

[0054] Step S102, construct an environmental stress factor model based on environmental data and preset weight coefficients.

[0055] Specifically, the environmental stress factor refers to various environmental factors and their corresponding coefficients that affect the battery performance in an environmental test.

[0056] Construct an environmental stress factor (ESF) model based on environmental data and preset weight coefficients to quantify the non-linear effects of environmental parameters such as temperature, humidity, salt mist concentration, and temperature change rate on battery aging.

[0057] Step S103, calculate the environmental stress factor value in real time based on the environmental stress factor model, and construct a mapping function between the environmental stress factor value and the battery health state at the corresponding timestamp based on the battery operation data.

[0058] Specifically, calculate the environmental stress factor ESF value, and establish a mapping function between the environmental stress factor ESF value and the battery SOH by using methods such as non-linear regression or neural network based on historical battery aging data.

[0059] Step S104, obtain the real-time battery SOH estimation value based on the battery operation data, the environmental stress factor value and the mapping function.

[0060] Specifically, input the real-time calculated ESF value and battery operation data into the mapping function to output the battery SOH estimation value in real time. To improve the accuracy of the battery SOH estimation value, use the extended Kalman filter or other dynamic correction algorithms to perform online filtering and adaptive correction on the input data and output results, and retrain the mapping function periodically.

[0061] The battery SOH estimation method provided in this embodiment can more comprehensively reflect the influence of the battery external environment by collecting environmental parameters such as temperature, humidity, salt mist concentration, and temperature change rate in real time, compared with the traditional single data collection method. Combining with the battery operation data, using the constructed Environmental Stress Factor (ESF) model and non-linear mapping function, it considers the online real-time estimation of the battery health state (SOH) under the influence of different environmental factors, effectively reflects the accelerating effect of the external environment on battery aging, improves the accuracy of the health state (SOH) estimation of the energy storage system under harsh climate conditions, extends the battery life and reduces the maintenance cost, and solves the problem of insufficient health state estimation of the energy storage system under harsh climate conditions.

[0062] In this embodiment, a battery SOH estimation method is provided, which can be used for an energy storage system. Figure 2 It is a flowchart of the battery SOH estimation method according to an embodiment of the present invention. As Figure 2 shown, the process includes the following steps:

[0063] Step S201, collect battery operation data and environmental parameters in real time. For details, please refer to Figure 1 step S101 of the embodiment shown, which will not be elaborated here.

[0064] Step S202, construct an environmental stress factor model based on the environmental data and preset weight coefficients.

[0065] Specifically, the environmental parameters include temperature, relative humidity, salt mist concentration, and temperature change rate; the preset weight coefficients include temperature weight coefficient, relative humidity weight coefficient, salt mist corrosion weight coefficient, and temperature change rate weight coefficient; the above step S202 includes:

[0066] Step S2021, calculate the product of the temperature and the temperature weight coefficient to obtain the temperature term, the product of the relative humidity and the relative humidity weight coefficient to obtain the humidity term, the product of the salt mist concentration and the salt mist corrosion weight coefficient to obtain the salt mist term, and the product of the temperature change rate and the temperature change rate weight coefficient to obtain the temperature change rate term. Among them, the temperature weight coefficient, relative humidity weight coefficient, salt mist corrosion weight coefficient, and temperature change rate weight coefficient are set in the following manner:

[0067] The battery is subjected to charge and discharge cycle tests using an orthogonal experimental design under different temperatures, different humidities, and different salt spray levels. When the battery capacity meets the preset capacity threshold, the battery capacity, internal resistance, and EIS spectrum during the charge and discharge cycle tests within the preset number of times are recorded, and the reciprocal of the number of cycles required for the capacity to decay to the preset capacity threshold is calculated as the battery aging rate. The proportion of the mean square error of the battery capacity, internal resistance, and EIS spectrum is calculated respectively as the weight benchmark, and the initial values of the temperature weight coefficient, relative humidity weight coefficient, salt spray corrosion weight coefficient, and temperature change rate weight coefficient are calculated based on the weight benchmark.

[0068] Step S2022: Construct an environmental stress factor model based on the temperature term, humidity term, salt spray term, and temperature change rate term, and perform offline training on the preliminary environmental stress factor model using offline calibration. Correct the preliminary environmental stress factor model according to the actual battery aging data to obtain the final environmental stress factor model. Specifically, the environmental stress factor (ESF) model is constructed through a non-linear regression algorithm, which can quantify the comprehensive impact of environmental factors on battery aging. The constructed environmental stress factor (ESF) model is as follows:

[0069]

[0070] In the formula, is the temperature term, is the humidity term, δ·ln(1 + C salt (t)) is the salt spray term, is the temperature change rate term. In the temperature term, α is the temperature weight coefficient, E a is the battery aging activation energy, calibrated through a constant current charge and discharge experiment, T(t) is the absolute temperature at time t (unit: Kelvin), k is the Boltzmann constant, and the temperature term mainly characterizes the influence of temperature on the aging rate. In the humidity term, β is the relative humidity weight coefficient, RH ref is the reference humidity, RH(t) is the relative humidity value at time t, γ is the humidity sensitivity index, and the humidity term mainly characterizes the influence of humidity on the aging rate. In the salt spray term, δ is the salt spray corrosion weight coefficient, C salt (t) is the salt spray concentration at time t, and the salt spray term mainly reflects the relationship between the salt spray concentration and the corrosion depth, that is, the influence of the salt spray concentration on battery aging. In the temperature change rate term, ε is the temperature mutation penalty factor, that is, the temperature change rate weight coefficient, and the temperature change rate term mainly characterizes that uneven internal stress in the battery caused by rapid temperature changes affects the battery life.

[0071] For the established ESF model, first calibrate the initial values of the weight coefficients α, β, and δ under different environmental scenarios. The specific experimental method adopts an orthogonal experimental design (Design of Experiments, DOE) and an accelerated aging test, aiming to quantify the temperature (T), humidity (RH), salt spray concentration (Csalt ) The independent and interactive effects on the SOH attenuation of the battery, and determine its initial weight in the environmental stress factor (ESF) model. The factor levels are selected as three different temperatures, three different humidities, and three different salt spray levels. For the three factors and three levels, an L9(3 4 ) orthogonal array is used, and a total of 9 groups of experiments are required. The batteries of the nine groups of experiments are subjected to charge and discharge cycle tests. A constant current charge and discharge of 1C is carried out, and the cut-off voltage is 3.0V - 4.2V. The cycle is repeated until the capacity decays to 80% SOH (termination condition). Record the number of cycles of the nine groups of experimental batteries from the start to reaching the termination condition, and record the capacity, internal resistance, and EIS (Electrochemical Impedance Spectroscopy, EIS) spectrum (0.1Hz - 10kHz) every 10 cycles.

[0072] After that, calculate the aging rate k of the nine groups of experimental batteries respectively. k is the reciprocal of the number of cycles required for the capacity to decay to 80% SOH (reaching the termination condition). Then calculate the Pearson correlation coefficients of each factor of capacity, internal resistance, and EIS spectrum with the k value respectively, and quantify the significance of the influence of each factor of capacity, internal resistance, and EIS spectrum on the k value. The calculation formula of the Pearson correlation coefficient is as follows:

[0073]

[0074] Among them, cov(X,Y) is the covariance of each factor and the k value, and σ X σ Y is the product of the standard deviations of each factor and the k value. The Pearson correlation coefficients ρ1, ρ2, and ρ3 of the three factors of capacity, internal resistance, and EIS spectrum with the k value are obtained respectively by calculation.

[0075] Finally, calculate the initial weight coefficient, and take the proportion of the Pearson correlation coefficient of each factor with the k value as the weight benchmark Finally, obtain the initial values of α, β, and δ.

[0076] Furthermore, after calculating the initial weight coefficient by taking the proportion of the Pearson correlation coefficient of each factor with the k value as the weight benchmark, a sensitivity analysis method can be further adopted to evaluate the sensitivity of each weight coefficient to the model output result. For the weight coefficients with high sensitivity, key optimization is carried out in the subsequent model training. In the optimization process, regularization techniques (such as L1, L2 regularization) are introduced to avoid the problem of model overfitting and improve the generalization ability of the model. By continuously adjusting the weight coefficients, the prediction error of the model on the training data set is minimized, while ensuring that the model has good performance on the test data set.

[0077] In the offline parameter calibration stage, the preliminary environmental stress factor model is trained offline using the non-linear least squares method or an optimization algorithm based on machine learning (such as particle swarm optimization, genetic algorithm, etc.). Based on the actually collected battery aging data (such as the data of the change of capacity, internal resistance, etc. with the number of cycles), by continuously adjusting the model parameters (each weight coefficient and other related parameters), the error between the environmental stress factor calculated by the model and the actual battery aging situation is minimized. During the training process, parameters such as appropriate iteration times and convergence conditions are set to ensure the effectiveness and stability of the optimization of the model parameters. Finally, the final environmental stress factor model that can accurately reflect the effect of environmental stress on battery aging is obtained.

[0078] Step S203, calculate the environmental stress factor value in real time based on the environmental stress factor model, and construct a mapping function between the environmental stress factor value and the battery health state corresponding to the time stamp based on the battery operation data. For details, please refer to Figure 1 Step S103 of the embodiment shown, which will not be elaborated here.

[0079] Step S204, obtain the real-time battery SOH estimation value based on the battery operation data, the environmental stress factor value and the mapping function. For details, please refer to Figure 1 Step S104 of the embodiment shown, which will not be elaborated here.

[0080] The battery SOH estimation method provided in this embodiment comprehensively incorporates environmental parameters such as temperature, relative humidity, salt fog concentration, and temperature change rate, and sets corresponding weight coefficients for each parameter. By calculating the product of each parameter and the weight coefficient respectively, independent temperature terms, humidity terms, salt fog terms, and temperature change rate terms are obtained. This method can accurately quantify the influence of different environmental factors on battery aging, better conform to the working conditions of the battery in a complex actual environment, avoid model deviation caused by incomplete consideration of environmental factors, and lay a foundation for accurately evaluating the degree of battery aging. Using offline calibration to train the preliminary environmental stress factor model offline and correcting the model based on the actual battery aging data can effectively improve the accuracy and reliability of the model. Through comparison and calibration with real data, the model can more realistically reflect the relationship between environmental factors and battery aging, and avoid estimation errors caused by inaccurate model parameters.

[0081] In this embodiment, a battery SOH estimation method is provided, which can be used in an energy storage system. Figure 3 is a flowchart of the battery SOH estimation method according to an embodiment of the present invention, as Figure 3 shown, and this process includes the following steps:

[0082] Step S301, collect battery operation data and environmental parameters in real time. For details, please refer to Figure 2 Step S201 of the embodiment shown, which will not be elaborated here.

[0083] Step S302: Construct an environmental stress factor model based on environmental data and preset weight coefficients. For details, please refer to Figure 2 Step S202 of the illustrated embodiment, which will not be elaborated herein.

[0084] Step S303: Calculate the environmental stress factor value in real time based on the environmental stress factor model, and construct a mapping function between the environmental stress factor value and the battery health state corresponding to the time stamp based on the battery operation data.

[0085] Specifically, the above-mentioned step S303 includes:

[0086] Step S3031: Substitute the environmental parameters into the environmental stress factor model to obtain the environmental stress factor value corresponding to the time stamp.

[0087] Specifically, substitute the preprocessed environmental parameters into the formula of each factor item of the environmental stress factor model respectively. For example, substitute the temperature item humidity item salt spray item δ·ln(1 + C salt (t)) and the temperature change rate item into formula (1) respectively to obtain the environmental stress factor ESF value corresponding to the time stamp. During the calculation process, a high-precision numerical calculation library is used to avoid result errors caused by insufficient calculation accuracy, and ensure that the ESF value can accurately reflect the environmental stress condition.

[0088] Step S3032: Calculate the battery capacity attenuation percentage consistent with the time stamp of the environmental stress factor value, and use the battery capacity attenuation percentage as the battery health state.

[0089] Specifically, after obtaining the battery capacity data consistent with the time stamp of the environmental stress factor value, perform in-depth analysis on the battery capacity data. First, use the Savitzky-Golay filtering algorithm to smooth the capacity data to eliminate the random noise in the capacity fluctuation and highlight the capacity attenuation trend; then, combine the initial capacity C0 of the battery to calculate the attenuation percentage of the battery capacity C(t) at each time stamp, and the formula is:

[0090] SOH capacity = C0 / C(t)×100% (3);

[0091] To more accurately reflect the battery health state, the influence of the depth of discharge (DOD) of the battery on the capacity attenuation can also be considered to correct the calculation result.

[0092] The calculated percentage of battery capacity attenuation is used as the core indicator of the battery health state. At the same time, in order to make the characterization of the battery health state more comprehensive, data such as battery internal resistance and EIS spectrum can be combined for comprehensive evaluation. For example, weights are set according to the growth of the internal resistance and weighted with the percentage of capacity attenuation to obtain the final battery health state value. In the fusion process, through the Principal Component Analysis (PCA) method, the weight coefficients of each indicator are determined, while reducing the data dimension, key information is retained, so that the battery health state can more truly reflect the actual aging degree of the battery.

[0093] Step S3033, based on the battery operation data, environmental stress factor values, battery health state, and a preset neural network, construct a mapping function between the environmental stress factor values and the battery health state at the corresponding timestamp.

[0094] Specifically, the obtained ESF values and the battery health state (SOH expressed as the percentage of capacity attenuation) at the corresponding moments are used as training samples, and data-driven methods such as nonlinear regression or a preset neural network are used to construct a mapping model, which can be formally expressed as:

[0095] SOH(t) = f(ESF(t), X(t)) (4);

[0096] Among them, X(t) includes other battery operation parameters (such as charge and discharge current, voltage, number of cycles, etc.) to make up for the possible deficiencies when only considering environmental factors. The mapping model can use typical methods such as Multilayer Perceptron (MLP), Long Short-Term Memory (LSTM), or Support Vector Regression (SVR).

[0097] Step S3034, introduce an adaptive weight and dynamic correction mechanism into the mapping function. When training the mapping function, use the mean square error as the loss function of the mapping function, and update the parameters of the mapping function through an adaptive optimization algorithm until the training error converges.

[0098] Specifically, to further improve the nonlinear fitting ability, an adaptive weight and dynamic correction mechanism can be introduced into the model. The mean square error (MSE) is used as the loss function for model training, and the parameters are updated through an adaptive optimization algorithm (such as Adam) until the training error converges, and finally a mapping model of ESF and SOH is obtained.

[0099] Furthermore, an adaptive weight mechanism is introduced into the neural network. By designing a special weight update formula, the network can dynamically adjust each connection weight according to the characteristics of the input data. For example, the attention mechanism (Attention Mechanism) is adopted to calculate the attention weights of different features in the input data, and the attention weights are multiplied by the original weights to achieve the adaptive adjustment of the weights, highlighting the factors that have a greater impact on the battery health state. The dynamic correction mechanism re-evaluates the model regularly (such as every 100 training iterations), and fine-tunes the model parameters using the newly collected data. Specifically, an incremental learning algorithm can be used to quickly adapt to the changes in new data while retaining the knowledge of the original model, ensuring that the model always maintains a high prediction accuracy during the battery aging process.

[0100] The mean square error (MSE) is used as the loss function of the mapping function, and an adaptive optimization algorithm (such as Adam, RMSProp, etc.) is used to update the parameters of the mapping function. Taking the Adam algorithm as an example, during the training process, the learning rate is dynamically adjusted according to the first-order moment estimate and the second-order moment estimate of the gradient, enabling the model to converge quickly in the initial stage of training and avoiding excessive oscillation when approaching the optimal solution. Appropriate hyperparameters, such as the initial value of the learning rate, are set, and the hyperparameters are optimized through the cross-validation method. At the same time, to prevent the model from overfitting, the Dropout technique is introduced, and neurons are randomly discarded with a certain probability during the training process to force the network to learn more robust feature representations until the training error converges, obtaining a stable and reliable mapping function between the environmental stress factor value and the battery health state.

[0101] Step S304, obtaining a real-time battery SOH estimation value based on the battery operation data, the environmental stress factor value, and the mapping function.

[0102] Specifically, the above step S304 includes:

[0103] Step S3041, inputting the battery operation data and the environmental stress factor value into the mapping function to obtain a preliminary battery SOH estimation value.

[0104] Specifically, the battery operation data and the environmental stress factor value are input into the mapping function (such as a neural network). Taking a multi-layer feedforward neural network as an example, the data enters from the input layer and is calculated successively through the neurons of each hidden layer. Each neuron processes the input signal through weighted summation and an activation function (such as ReLU, Sigmoid, etc.), and the information is passed between the hidden layers through the learned weight parameters, and finally a preliminary battery SOH estimation value is obtained at the output layer. During the calculation process, the GPU acceleration technology is used to improve the data processing and calculation efficiency, ensuring that a large amount of data can be predicted in a short time to meet the real-time requirements.

[0105] Step S3042: Perform online filtering and adaptive correction on the preliminary estimated value of the battery SOH to obtain the real-time estimated value of the battery SOH.

[0106] Specifically, the extended Kalman filter (EKF) or other suitable online filtering algorithms are used to process the preliminary estimated value of the battery SOH. Taking EKF as an example, its core idea is to predict and update the battery SOH state through the state equation and the observation equation. First, based on the SOH estimated value at the previous moment and the system dynamic model (considering the influence of battery operating parameters and environmental factors), predict the SOH state at the current moment; then, take the preliminary estimated value as the observation data, combine it with the predicted value, and correct the predicted value through the calculation of the Kalman gain to obtain the filtered SOH estimated value. During the filtering process, the covariance matrix is updated in real time to adapt to the changes in the battery state and the uncertainty of the measurement noise, effectively eliminating the random noise in the data and making the estimated value smoother and more stable.

[0107] The adaptive correction mechanism dynamically adjusts the model parameters by comparing the filtered SOH estimated value with the actual measurement data (such as the battery capacity, internal resistance, etc. collected again). Using the error feedback principle, calculate the error between the estimated value and the actual value, and according to the magnitude and trend of the error, use an adaptive optimization algorithm (such as a variant algorithm of stochastic gradient descent) to fine-tune the parameters of the mapping function. For example, when the error is large, increase the learning rate to accelerate the parameter update speed; when the error is small, decrease the learning rate to avoid over-adjustment. At the same time, combined with the dynamic changes during the battery aging process, regularly optimize the parameters of the adaptive correction (such as the learning rate adjustment strategy, error threshold, etc.) to ensure that the correction process can adapt to the changes in the battery state in a timely and accurate manner, making the finally obtained real-time estimated value of the battery SOH closer to the actual health state of the battery.

[0108] The battery SOH estimation method provided in this embodiment uses the percentage of battery capacity attenuation consistent with the time stamp of the environmental stress factor value as the battery health state, accurately corresponding to the real-time health condition of the battery under specific environmental stress. A mapping function is constructed based on battery operation data, environmental stress factor values, and battery health state, integrating multi-source information such as the battery's own operation parameters, external environmental influence factors, and actual health conditions. A preset neural network is used to construct the mapping function. The powerful non-linear fitting ability of the neural network can automatically learn and extract complex features and rules in the data, effectively processing the highly non-linear mapping relationship between the environmental stress factor value and the battery health state. Emphasizing the consistency of the environmental stress factor value and the battery health state time stamp ensures the accuracy of the logical relationship between the data. An adaptive weight mechanism is introduced, enabling the mapping function to automatically adjust the influence weights of various factors on the battery health state assessment according to the changes in the input data. The addition of a dynamic correction mechanism allows the mapping function to continuously optimize itself as the battery ages and the environment changes. The mean square error is used as the loss function, which can intuitively measure the difference between the model prediction value and the actual battery health state, providing a clear target orientation for model optimization. The adaptive weight, dynamic correction mechanism, mean square error loss function, and adaptive optimization algorithm cooperate with each other to form an organic whole, comprehensively enhancing the comprehensive performance of the mapping function, enabling it to stably and accurately realize the mapping between the environmental stress factor value and the battery health state in complex environments and the dynamic process of battery aging.

[0109] As one or more specific application embodiments of the embodiment of the present invention, in combination with Figure 4 the battery SOH estimation method provided by the present invention is further described in detail. As Figure 4 shown, the specific process is as follows:

[0110] S1: Real-time collect battery operation data and environmental parameters through a weather-resistant sensor network.

[0111] Specifically, a weather-resistant sensor network is arranged in the energy storage system to collect battery operation data (such as voltage, current, charge and discharge cycle times, etc.) and environmental parameters (such as temperature, relative humidity, salt fog concentration, temperature change rate, etc.) in real time. The weather-resistant sensor network includes a temperature and humidity sensor, a salt fog concentration sensor, a voltage sensor, and a current sensor.

[0112] S2: Perform data preprocessing operations such as data cleaning, noise filtering, interpolation completion, and normalization on the collected battery operation data and environmental parameters to ensure the accuracy and stability of the input data.

[0113] S3: Construct an environmental stress factor (ESF) model to quantify the non-linear effects of temperature, humidity, and salt fog concentration on battery aging.

[0114] Specifically, by using a non-linear regression algorithm to construct an environmental stress factor model, the comprehensive impact of environmental factors on battery aging can be quantified. The constructed Environmental Stress Factor (ESF) model is as follows:

[0115]

[0116] In the formula, is the temperature term, is the humidity term, δ·ln(1 + C salt (t)) is the salt spray term, is the temperature change rate term. In the temperature term, α is the temperature weight coefficient, E a is the battery aging activation energy, calibrated through constant current charge and discharge experiments, T(t) is the absolute temperature at time t (unit: Kelvin), k is the Boltzmann constant, and the temperature term mainly characterizes the impact of temperature on the aging rate. In the humidity term, β is the relative humidity weight coefficient, RH ref is the reference humidity, RH(t) is the relative humidity value at time t, and γ is the humidity sensitivity index. The humidity term mainly characterizes the impact of humidity on the aging rate. In the salt spray term, δ is the salt spray corrosion weight coefficient, C salt (t) is the salt spray concentration at time t, and the salt spray term mainly reflects the relationship between the salt spray concentration and the corrosion depth, that is, the impact of the salt spray concentration on battery aging. In the temperature change rate term, ε is the temperature mutation penalty factor, that is, the temperature change rate weight coefficient, and the temperature change rate term mainly characterizes that under rapid temperature changes, the internal stress of the battery is uneven, thus affecting the battery life.

[0117] For the established ESF model, first, the initial values of the weight coefficients α, β, and δ in different environmental scenarios are calibrated. The specific experimental method adopts the Orthogonal Experiment Design (DOE) and accelerated aging tests, aiming to quantify the independent and interactive effects of temperature (T), humidity (RH), and salt spray concentration (C salt ) on the attenuation of the battery's State of Health (SOH), and determine their initial weights in the Environmental Stress Factor (ESF) model. The factor levels are selected as three different temperatures, three different humidities, and three different salt spray degrees. For the three factors and three levels, the L9(3 4 ) orthogonal table is selected, and a total of 9 groups of experiments are required. The charge and discharge cycle tests are carried out on the nine groups of experimental batteries with a constant current charge and discharge of 1C, and the cut-off voltage is 3.0V - 4.2V. The cycle is carried out until the capacity decays to 80% of the SOH (termination condition). Record the number of cycles of the nine groups of experimental batteries from the start to reaching the termination condition, and record the capacity, internal resistance, and Electrochemical Impedance Spectroscopy (EIS) spectrum (0.1Hz - 10kHz) every 10 cycles.

[0118] After that, the aging rates k of nine groups of experimental batteries are calculated respectively. k is the reciprocal of the number of cycles required for the capacity to decay to 80% SOH (reaching the termination condition). Then, the Pearson correlation coefficients of each factor of capacity, internal resistance, and EIS spectrum with the k value are calculated respectively to quantify the significance of the influence of each factor of capacity, internal resistance, and EIS spectrum on the k value. The calculation formula of the Pearson correlation coefficient is as follows:

[0119]

[0120] where cov(X,Y) is the covariance of each factor and the k value, and σ X σ Y is the product of the standard deviations of each factor and the k value. By calculation, the Pearson correlation coefficients ρ1, ρ2, and ρ3 of the three factors of capacity, internal resistance, and EIS spectrum with the k value are obtained respectively.

[0121] Finally, the initial weight coefficients are calculated, and the proportion of the Pearson correlation coefficient of each factor with the k value is used as the weight basis Finally, the initial values of α, β, and δ are obtained.

[0122] Furthermore, after calculating the initial weight coefficients with the proportion of the Pearson correlation coefficient of each factor with the k value as the weight basis, a sensitivity analysis method can be further used to evaluate the sensitivity of each weight coefficient to the model output result. For the weight coefficients with high sensitivity, key optimization is carried out in the subsequent model training. In the optimization process, regularization techniques (such as L1 and L2 regularization) are introduced to avoid the problem of model overfitting and improve the generalization ability of the model. By continuously adjusting the weight coefficients, the prediction error of the model on the training data set is minimized, while ensuring that the model has good performance on the test data set.

[0123] In the offline parameter calibration stage, the preliminary environmental stress factor model is offline trained using the nonlinear least squares method or machine learning-based optimization algorithms (such as particle swarm optimization, genetic algorithm, etc.). Based on the actual collected battery aging data (such as the change data of capacity, internal resistance, etc. with the number of cycles), by continuously adjusting the model parameters (each weight coefficient and other related parameters), the error between the environmental stress factor calculated by the model and the actual battery aging situation is minimized. During the training process, parameters such as appropriate iteration times and convergence conditions are set to ensure the effectiveness and stability of the model parameter optimization, and finally, the final environmental stress factor model that can accurately reflect the effect of environmental stress on battery aging is obtained.

[0124] S4: Using historical battery aging data, a mapping function between ESF and battery SOH is established by methods such as nonlinear regression or neural network, which can be formally expressed as:

[0125] SOH(t) = f(ESF(t), X(t)) (4);

[0126] Among them, X(t) includes other battery operating parameters (such as charge and discharge current, voltage, number of cycles, etc.) to make up for possible deficiencies when only considering environmental factors. The mapping model can adopt typical multi-layer feedforward neural network (MLP), long short-term memory network (LSTM), or methods such as support vector regression (SVR).

[0127] S5: Input the real-time calculated ESF value and battery operating data into the mapping function, and output the battery SOH estimated value in real time.

[0128] Specifically, according to the real-time input battery operating data, after preprocessing by the data preprocessing module, combined with the ESF model calculation to obtain the real-time ESF value, and input it into the trained mapping function to obtain the SOH estimated value of the current battery.

[0129] S6: Use the extended Kalman filter or other dynamic correction algorithms to perform online filtering and adaptive correction on the input data and output results, and retrain the mapping function periodically.

[0130] Specifically, to eliminate the random noise in the environmental data, an extended Kalman filter (EKF) is embedded in the system, and the filter is used to dynamically correct the ESF input and model output, so that the SOH estimation result is smoother and responds to the real state changes. The system should also retrain the mapping model periodically (for example, every 100 cycles or every day) to capture the tiny changes in the battery aging process and ensure the estimation accuracy during long-term use.

[0131] The battery SOH estimation method provided in this embodiment, by collecting environmental parameters such as temperature, humidity, salt mist concentration and temperature change rate in real time, combining with battery operating data, using the environmental stress factor (ESF) model and non-linear mapping function obtained by offline calibration, considers the online real-time estimation of the battery health state (SOH) under the influence of different environmental factors, effectively reflects the accelerating effect of the external environment on battery aging, and at the same time supports online adaptive correction, thereby optimizing the charge and discharge strategy, extending the battery life and reducing the maintenance cost.

[0132] In this embodiment, a battery SOH estimation device is also provided. This device is used to implement the above-mentioned embodiment and preferred implementation manners, and those that have been described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that realizes a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0133] This embodiment provides a battery SOH estimation device, as Figure 5 shown, including:

[0134] The data acquisition module 501 is used to collect battery operation data and environmental parameters in real time.

[0135] The environmental stress factor model construction module 502 is used to construct an environmental stress factor model based on environmental data and preset weight coefficients.

[0136] The mapping function construction module 503 is used to calculate the environmental stress factor value in real time based on the environmental stress factor model, and construct a mapping function between the environmental stress factor value and the battery health state at the corresponding timestamp based on the battery operation data.

[0137] The real-time battery SOH estimation value calculation module 504 is used to obtain the real-time battery SOH estimation value based on the battery operation data, the environmental stress factor value, and the mapping function.

[0138] In some alternative embodiments, the environmental parameters include temperature, relative humidity, salt fog concentration, and temperature change rate; the preset weight coefficients include a temperature weight coefficient, a relative humidity weight coefficient, a salt fog corrosion weight coefficient, and a temperature change rate weight coefficient; the environmental stress factor model construction module 502 includes:

[0139] The product calculation unit is used to calculate the product of temperature and the temperature weight coefficient to obtain the temperature term, the product of relative humidity and the relative humidity weight coefficient to obtain the humidity term, the product of salt fog concentration and the salt fog corrosion weight coefficient to obtain the salt fog term, and the product of temperature change rate and the temperature change rate weight coefficient to obtain the temperature change rate term.

[0140] The environmental stress factor model construction unit is used to construct an environmental stress factor model based on the temperature term, the humidity term, the salt fog term, and the temperature change rate term, and perform offline training on the preliminary environmental stress factor model using offline calibration, and correct the preliminary environmental stress factor model according to the actual battery aging data to obtain the final environmental stress factor model. The above temperature weight coefficient, relative humidity weight coefficient, and salt fog corrosion weight coefficient are set in the following manner:

[0141] Orthogonal experimental design is adopted to conduct charge and discharge cycle tests on the battery under different temperatures, different humidities, and different salt fog degrees; when the battery capacity meets the preset capacity threshold, record the battery capacity, internal resistance, and EIS spectrum during the charge and discharge cycle tests within the preset number of times, and calculate the reciprocal of the number of cycles required for the capacity to decay to the preset capacity threshold as the battery aging rate; calculate the proportion of the mean square error of the battery capacity, internal resistance, and EIS spectrum respectively as the weight reference, and calculate the initial values of the temperature weight coefficient, relative humidity weight coefficient, and salt fog corrosion weight coefficient based on the weight reference.

[0142] In some alternative embodiments, the mapping function construction module 503 includes:

[0143] Substitute the environmental parameters into the environmental stress factor model to obtain the environmental stress factor value at the corresponding timestamp. The battery health state calculation unit is used to calculate the battery capacity attenuation percentage that is consistent with the timestamp of the environmental stress factor value, and use the battery capacity attenuation percentage as the battery health state.

[0144] The mapping function construction unit is used to construct a mapping function between the environmental stress factor value and the battery health state at the corresponding timestamp based on the battery operation data, the environmental stress factor value, the battery health state, and a preset neural network.

[0145] The mapping function optimization unit is used to introduce an adaptive weight and a dynamic correction mechanism into the mapping function. When training the mapping function, the mean square error is used as the loss function of the mapping function, and the parameters of the mapping function are updated through an adaptive optimization algorithm until the training error converges.

[0146] In some optional embodiments, the real-time battery SOH estimation module 504 includes:

[0147] The battery SOH preliminary estimation unit is used to input the battery operation data and the environmental stress factor value into the mapping function to obtain the battery SOH preliminary estimation value.

[0148] The filtering unit is used to perform online filtering and adaptive correction on the battery SOH preliminary estimation value to obtain the real-time battery SOH estimation value.

[0149] The further function descriptions of the above-mentioned various modules and units are the same as those in the corresponding embodiments above, and will not be elaborated here.

[0150] The battery SOH estimation device in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0151] The embodiment of the present invention also provides a computer device having the above-mentioned Figure 5 shown battery SOH estimation device.

[0152] Please refer to Figure 6 , Figure 6 which is a schematic structural diagram of a computer device provided by an optional embodiment of the present invention. As shown in Figure 6As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 6 Take one processor 10 as an example in

[0153] The processor 10 can be a central processing unit, a network processor, or a combination thereof. Among them, the processor 10 can further include a hardware chip. The above hardware chip can be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The above programmable logic device can be a complex programmable logic device, a field programmable gate array, a generic array logic, or any combination thereof.

[0154] Among them, the memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiments.

[0155] The memory 20 can include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the computer device. In addition, the memory 20 can include high-speed random access memory, and can also include non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 can optionally include a memory remotely set relative to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0156] The memory 20 can include volatile memory, such as random access memory; the memory can also include non-volatile memory, such as flash memory, a hard disk, or a solid-state drive; the memory 20 can also include a combination of the above types of memory.

[0157] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 may be connected through a bus or other means. Figure 6 Taking the connection through the bus as an example.

[0158] The input device 30 can receive input digital or character information, and generate key signal inputs related to the user settings and function controls of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED), and a haptic feedback device (e.g., a vibration motor), etc. The above display device includes, but is not limited to, a liquid crystal display, a light-emitting diode, a display, and a plasma display. In some alternative embodiments, the display device may be a touch screen.

[0159] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiment is implemented.

[0160] A part of the present invention can be applied as a computer program product, such as computer program instructions. When executed by a computer, through the operation of the computer, the methods and / or technical solutions according to the present invention can be invoked or provided. Those skilled in the art should be able to understand that the forms of existence of computer program instructions in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible by the computer.

[0161] Although embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for estimating the state of health (SOH) of a battery, characterized in that, The method includes: Collecting battery operation data and environmental parameters in real time; Constructing an environmental stress factor model based on environmental data and preset weight coefficients; Calculating the environmental stress factor value in real time based on the environmental stress factor model, and constructing a mapping function between the environmental stress factor value and the battery health state at the corresponding timestamp based on the battery operation data; Obtaining the real-time battery SOH estimation value based on the battery operation data, environmental stress factor value, and mapping function.

2. The method according to claim 1, characterized in that, The environmental parameters include temperature, relative humidity, salt fog concentration, and temperature change rate; the preset weight coefficients include temperature weight coefficient, relative humidity weight coefficient, salt fog corrosion weight coefficient, and temperature change rate weight coefficient; The constructing of the environmental stress factor model based on the battery operation data, environmental data, and preset weight coefficients includes: Calculating the product of temperature and temperature weight coefficient to obtain the temperature term, the product of relative humidity and relative humidity weight coefficient to obtain the humidity term, the product of salt fog concentration and salt fog corrosion weight coefficient to obtain the salt fog term, and the product of temperature change rate and temperature change rate weight coefficient to obtain the temperature change rate term respectively; Constructing a preliminary environmental stress factor model based on the temperature term, humidity term, salt fog term, and temperature change rate term, and performing offline training on the preliminary environmental stress factor model by using offline calibration, and correcting the preliminary environmental stress factor model according to the actual battery aging data to obtain the final environmental stress factor model.

3. The method according to claim 2, wherein The temperature weight coefficient, relative humidity weight coefficient, and salt fog corrosion weight coefficient are set in the following manner: Performing charge and discharge cycle tests on the battery by using orthogonal experimental design at different temperatures, different humidities, and different salt fog degrees; When the battery capacity meets the preset capacity threshold, recording the battery capacity, internal resistance, and EIS spectrum in the charge and discharge cycle tests within the preset number of times, and calculating the reciprocal of the number of cycles required for the capacity to decay to the preset capacity threshold as the battery aging rate; Calculating the proportion of the mean square error of the battery capacity, internal resistance, and EIS spectrum respectively as the weight benchmark, and calculating the initial values of the temperature weight coefficient, relative humidity weight coefficient, and salt fog corrosion weight coefficient based on the weight benchmark.

4. The method according to claim 1, characterized in that The calculating of the environmental stress factor value in real time based on the environmental stress factor model includes: Substituting the environmental parameters into the environmental stress factor model to obtain the environmental stress factor value at the corresponding timestamp.

5. The method according to claim 1, wherein The constructing of the mapping function between the environmental stress factor value and the battery health state at the corresponding timestamp based on the battery operation data includes: Calculating the percentage of battery capacity decay consistent with the timestamp of the environmental stress factor value, and taking the percentage of battery capacity decay as the battery health state; Constructing the mapping function between the environmental stress factor value and the battery health state at the corresponding timestamp based on the battery operation data, environmental stress factor value, battery health state, and preset neural network.

6. The method according to claim 5, wherein The constructing of the mapping function between the environmental stress factor value and the battery health state at the corresponding timestamp based on the battery operation data further includes: An adaptive weight and a dynamic correction mechanism are introduced into the mapping function. When training the mapping function, the mean square error is used as the loss function of the mapping function, and the parameters of the mapping function are updated through an adaptive optimization algorithm until the training error converges.

7. The method according to claim 1, wherein The obtaining of the real-time battery SOH estimation value based on the battery operation data, the environmental stress factor value, and the mapping function includes: Inputting the battery operation data and the environmental stress factor value into the mapping function to obtain a preliminary battery SOH estimation value; Performing online filtering and adaptive correction on the preliminary battery SOH estimation value to obtain the real-time battery SOH estimation value.

8. A battery SOH estimation device, characterized in that, The device includes: A data acquisition module, configured to collect battery operation data and environmental parameters in real time; An environmental stress factor model construction module, configured to construct an environmental stress factor model based on environmental data and a preset weight coefficient; A mapping function construction module, configured to calculate the environmental stress factor value in real time based on the environmental stress factor model, and construct a mapping function between the environmental stress factor value and the battery health state corresponding to the time stamp based on the battery operation data; A real-time battery SOH estimation value calculation module, configured to obtain a real-time battery SOH estimation value based on the battery operation data, the environmental stress factor value, and the mapping function.

9. A computer device, characterized in that, including: A memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to execute the battery SOH estimation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Computer instructions are stored on the computer-readable storage medium, and the computer instructions are used to cause a computer to execute the battery SOH estimation method according to any one of claims 1 to 7.

Images